Reading disability (RD) is a serious, life-long condition that has significant public health consequences for society. Past NIH research has provided a lot of information about reading development and intervention in the early grades (K-3). However, few studies have investigated the nature of children who develop RD in the intermediate grades (4th-5th grade). Estimates of this RD (coined by Leach et al. as Late-Emerging RD, or LERD) suggest that the prevalence of LERD is non-trivial, ranging from 20% to 46% of those identified with RD in late elementary school. Thus, it is likely that LERD is a significant contributing factor to the substantial prevalence of reading failure in 4th and 8th grades (which is about 30%; NAEP, 2007) and suggests an urgent public health need for understanding more about the underlying causes of LERD. Within this context, the goal of the current project is to determine the cognitive and neurobiological profile associated with LERD, and establish how LERD is similar or different than early reading failure (RD-Early, or RD-E). To accomplish this goal, the proposed study is composed of two aims. Specific Aim 1 is to identify concurrent behavioral and neuronal weaknesses in a sample of 5th grade children with LERD. The plan is to capitalize on a unique situation in which a randomly selected longitudinal cohort that is part of a previous study (not originally designed to study LERD) is available to us. We hypothesize that children with LERD will show a distinct cognitive and neurobiological profile compared to RD-E, with particular weaknesses in various aspects of oral language and executive function. Specific Aim 2 of the study is to use the areas of identified weaknesses in the fifth graders to guide the development of a predictive battery of cognitive and neurobiological measures to characterize the behavioral and neurobiological profiles of those at risk for LERD in earlier grades. We hypothesize that oral language deficits and executive function deficits will feature prominently in LERD. We plan to follow this second cohort of children for several years, acquiring behavioral and neurobiological data to better understand the development of LERD. Ultimately, this line of research seeks to change current practice by determining what measures schools should include as early identifiers for those at risk for LERD, as well as develop early intervention programs for those at risk for LERD. In addition, by use of both neurobiological and cognitive measures, we will learn not only specifically about LERD, but also will contribute generally to understanding brain maturation, particularly connectivity, over time. Thus, this research will not only be valuable to understanding LERD, but also will be more generally helpful to understanding child development, as it will serve to provide information about brain maturation and its relationship to cognition in developing children.